Researchers from the Fred Hutchinson Cancer Research Center have developed a droplet digital PCR assay to precisely enumerate a class of T-cells called tumor-infiltrating lymphocytes (TILs) from cancer tissue, and used the method to demonstrate that the presence and quantity of these cells are strongly correlated with increased survival in ovarian cancer patients.
The scientists believe that the assay can be extrapolated to different cancer types and eventually prove useful as a clinical tool to aid prognosis or predict how patients will respond to immunotherapies — an application that Fred Hutchinson spinout Adaptive Biotechnologies is now pursuing.
The research, which was published this week in Science Translational Medicine, also demonstrates the potential of droplet digital PCR as a tool for clinical applications requiring absolute quantification of nucleic acid targets.
"There's no way you could do this with any method other than digital PCR because of the numerous primer pairs and probes that we have," Jason Bielas, a researcher at the Hutch and corresponding author on the study, said in a statement. "Digital PCR partitions all the reactions so you can amplify these targets independently of PCR efficiency without any competing side reactions."
The assay, which the group calls QuanTILfy, grew out of Bielas' general research interest — accurately detecting and quantifying mutations in heterogeneous tumor tissues for the purposes of identifying potential biomarkers.
"When we analyze tumors … it is not just one cell type," Bielas told PCR Insider this week. "When we extract DNA, the assumption a lot of the time is that you're analyzing from one cell type. But it was pretty clear from my mutation work that I didn't know exactly where I was measuring mutations from."
This idea extends to the way immune cells in the body attack tumor cells. Several other research groups have shown that TILs directly attack tumor cells in a variety of cancer types, and that the presence of these immune cells is correlated with patient survival. However, T-cell receptors undergo various rearrangements to address different challenges and then expand in a process called clonality. Furthermore, tumors are heterogeneous in nature and may harbor mutational targets for TILs in some cells but not others, even in the same tumor.
Given all of these variables, no reliable method had yet been developed to accurately measure the extent of these TIL invasions in tumor tissues.
"By looking at a tumor [using immunohistochemistry] you can see there are T cells that infiltrate there, and so the question becomes: Is there a way that we can quantify this quickly?" Bielas said.
Harlan Robins, one of Bielas' colleagues at the Hutch and first author on the paper, had previously developed technology to effectively sequence the gene rearrangements responsible for clonality by correcting for multiplex PCR amplification bias — work that he and colleagues described in a study published in October and which Adaptive is using in its immune cell sequencing services.
"We were talking and I said that we can count exactly how many T-cells there are if we combine our technologies," Bielas said. "They designed a primer set … to count the number of recombined T-cell receptors, which we've optimized for digital PCR. We designed a bunch of TaqMan probes with that with the idea that now we can count exactly how many T-cells are in a tumor."
Specifically, Bielas and colleagues designed a multiplexed droplet digital PCR assay to amplify rearranged T-cell receptor β loci from genomic DNA using the Bio-Rad QX100 system. The assay comprised 45 forward primers, each specific to one or multiple functional T-cell receptor variable β gene segments; and 13 reverse primers each specific to a T-cell receptor joining β segment. The multiplex reaction also contained one of a series of 35 TaqMan probes complementary to 52 different variable β gene segments.
They further modified the assay to increase throughput, effectively creating an eight-well, multiplex ddPCR assay in which each T-cell receptor β gene was measured once in each of the wells. The sum of the counts from all eight wells then yielded a precise digital count of the total number of rearranged T-cell receptor β genes in the sample, and also allowed detection of one or more large clonal expansions.
The researchers performed the QuanTILfy assay on frozen primary tumor samples from 30 ovarian carcinoma patients with known survival outcomes, finding that on average TIL frequency was about threefold higher in patients who survived more than five years as compared with patients who survived less than two years.
"We know that T cells are involved in fighting the tumor in some way, and there are a number of drugs coming on the market that have been shown to interact positively with … T-cell activation," Bielas said. "One would expect that this test, because it's reproducible and quantitative, could find its way into the clinic, and maybe even act as a predictive biomarker for immunotherapy — determine patients that respond differently to different types of therapy … and stratify patient groups to get different types of treatments based on their TIL counts."
Bielas' team also demonstrated that QuanTILfy can be used to accurately and reproducibly characterize T-cell clonality in patients with T-cell acute lymphoblastic leukemia — results that they validated using deep sequencing.
"T-cell leukemia shows a single clone that gets expanded, and we picked that out in a lot of control samples," Bielas said. "These are not called tumor-infiltrating lymphocytes [in this case], but nevertheless it's validation that if there were a clonal expansion, we could pick it up."
Although the research was conducted under the auspices of the Hutch and various private and governmental funding agencies, the Hutch and Adaptive Biotech have filed for a joint patent on the method. Bielas and Robins both have a financial interest in Adaptive Biotech, and Robins consults for the company.
Should the assay move forward as a clinical tool, Adaptive will likely serve as the commercializing entity.
"Adaptive has a strong interest in commercializing measurements of tumor infiltrating lymphocytes for improved staging of tumors and for predictive biomarkers of response to immunotherapies," Robins told PCR Insider in an email, adding that the company "is working to refine our technologies and develop clinical data sets."
But it remains to be seen whether droplet digital PCR, as a technology platform, is ready for clinical implementation. Bielas said that QuanTILfy is a "first step" toward moving droplet digital PCR into the clinic.
"This assay is going to provide information that wasn't previously available, and it does it with use of [Bio-Rad's] system," he said. "Now, whether that system can be packaged for the clinic is something that I hope they are going to do. Clearly there is going to have to be a lot of modification to the equipment to get it into the clinic, which is something I'm sure [Bio-Rad] must be working on — they really know what's required."